Business Internet Redundancy and Failover: Keeping Your Business Online When the Connection Fails

Every Australian business that relies on the internet to operate faces the same risk: a single connection is a single point of failure. When that connection drops — whether for five minutes or five hours — the impact is immediate and measurable. Payment systems go offline. Cloud software becomes inaccessible. VoIP phones go silent. Remote staff lose access. In a business environment where nearly every revenue-generating activity depends on internet connectivity, unplanned downtime is not a minor inconvenience. It is a direct cost.

This article is for Australian SMB owners who want to understand what internet redundancy means in practice, what the available options are, and how to choose an approach that is proportionate to their risk and budget. Whether you have experienced a costly outage already or you are building proactively, the principles are the same.

The Single Point of Failure Problem

The phrase "single point of failure" comes from engineering, but it applies perfectly to business internet infrastructure. A single point of failure is any component whose failure alone is enough to bring down a system. For most Australian SMBs, the NBN connection — and the equipment connecting to it — is exactly that.

Consider the ways a single NBN connection can fail.

A physical cable fault on the local network can take down a building or a street. These faults are more common than most business owners assume, and NBN Co's standard resolution time for a business fault ranges from hours to days depending on the fault type and the service level of your plan.

A provider outage in your area — caused by equipment failure, software faults, or maintenance windows — can affect hundreds or thousands of services simultaneously. You have no control over when it occurs or how long it lasts.

Your premises equipment — the NBN connection device, your router, or cabling internal to your building — can fail without any fault on the network side. A power surge, a hardware failure, or a misconfiguration after a firmware update can all render a functioning NBN service useless.

Provider maintenance windows, while often scheduled, are not always communicated clearly or timed conveniently for business operations.

In each of these scenarios, if you have a single connection and it fails, you are offline. There is no debate, no workaround, no manual backup — you simply stop.

The question for every business owner is not whether this will happen, but what it will cost when it does, and whether the cost of prevention is lower than the cost of the outage.

For most Australian SMBs operating in the cloud — using cloud accounting, CRM, VoIP, point-of-sale, remote access, or any combination — the cost of even a two-hour outage during business hours outweighs the ongoing cost of a basic redundancy solution many times over.

What Internet Redundancy Actually Means

Internet redundancy, at its most straightforward, means having a second internet connection that activates when the first one fails. The key word is "activates" — this is automatic failover, not a manual switchover that requires a staff member to notice the outage, locate the secondary hardware, and manually change the network configuration.

A true redundancy setup has three components working together.

The first is a secondary internet connection that is active, tested, and ready. It should use a different network from your primary connection — ideally a different technology entirely. If your primary is NBN fixed line, your secondary should be 4G or 5G mobile broadband, or at minimum a fixed wireless connection from a different provider. The reason is straightforward: if a physical event (such as a cable cut or exchange equipment failure) takes out your primary connection, you want to ensure that the same physical event cannot also take out your secondary. Two NBN connections from different providers in the same area will often share the same physical infrastructure at some point, meaning a single fault can disable both.

The second component is a router capable of managing two active WAN connections simultaneously. A standard residential router or a basic business router typically supports only one WAN connection. A dual WAN router actively monitors both connections and manages the transition when one fails.

The third component is automatic failover logic — the configuration within the router that defines what constitutes a failure, how quickly the router detects it, and how traffic is redirected when failover occurs. A well-configured failover should detect connection loss within 30 to 60 seconds and redirect traffic to the secondary connection without requiring any staff intervention.

Without all three components, what you have is not redundancy — it is a contingency plan that requires manual execution under pressure, during an outage, when staff have other priorities. That is a very different thing.

The Four Redundancy Options — From Simple to Comprehensive

There is no single redundancy solution that suits every business. The right option depends on the volume of traffic your secondary connection needs to handle, how quickly you need failover to occur, your budget, and the complexity of your network environment. These four options cover the range from simple and affordable to enterprise-grade.

4G/5G Failover Router

A 4G failover router is the most accessible and widely appropriate redundancy solution for Australian SMBs. The router has a built-in or attached modem that accepts a 4G or 5G SIM card. When the primary NBN connection fails — detected through automatic ping or DNS monitoring — the router immediately routes all traffic over the mobile connection. When the primary connection restores, traffic switches back automatically.

Hardware cost is typically in the range of $300 to $800 for a quality business-grade router from manufacturers such as Draytek, Peplink, or Cradlepoint, depending on the model and capability. Ongoing cost is $30 to $80 per month for a business data SIM with a suitable data allowance.

The main limitation of 5G backup internet and 4G failover is bandwidth. Mobile connections in most Australian metropolitan and regional areas can deliver between 20 and 100 Mbps, which is adequate for essential business functions — VoIP, email, cloud software, payment terminals, remote access — but is not appropriate as the primary connection for businesses with high bandwidth demands such as large file transfers or video production workflows. As a failover for essential operations, it is exactly right.

This is the starting point for most SMBs because the cost is low, the technology is reliable, and 4G coverage in Australian metropolitan and suburban areas is now mature and consistent.

Dual WAN with a Second Fixed Connection

A step up from mobile failover is connecting two fixed-line services to a dual WAN router. The primary might be an NBN HFC or FTTP service, and the secondary could be a fixed wireless connection, a different NBN technology type, or a fibre service from a second provider.

The advantage over 4G failover is speed on the failover connection. If your secondary is another fixed broadband service, failover performance can be close to your primary, meaning business operations continue with minimal visible impact. Some dual WAN routers also support load balancing — distributing traffic across both connections simultaneously during normal operation, which improves aggregate throughput.

The trade-off is cost and complexity. Running two fixed-line services at your premises means two monthly service fees, potentially two separate contracts, and the need for both services to be provisioned and active. Total monthly cost for a dual fixed WAN setup could be $150 to $400 depending on the services chosen. It is appropriate for businesses where the secondary connection needs to carry significant traffic loads during failover, such as those with many simultaneous users or video conferencing intensive workflows.

SD-WAN with Multiple Connections

Software-defined wide area network (SD-WAN) technology moves beyond simple failover into intelligent traffic management across multiple simultaneous connections. Rather than designating one connection as primary and another as secondary, SD-WAN actively monitors the quality of all available connections and routes different types of traffic over whichever connection is most appropriate at any given moment.

VoIP traffic might be prioritised over the lowest latency connection. Large file transfers might be distributed across multiple connections to maximise throughput. If one connection degrades or fails, the SD-WAN platform automatically redistributes traffic without interruption.

SD-WAN is appropriate for larger businesses, multi-site operations, or organisations with complex application requirements and the IT management capacity to configure and maintain the platform. It is not typically the starting point for a single-site SMB with fewer than 50 staff.

Dedicated Internet with Carrier-Grade SLA

A different approach to resilience is addressing the problem at the primary connection level rather than adding a secondary. Enterprise ethernet and dedicated internet access products come with contractual service level agreements that specify maximum restoration times — often four hours or less for a full service outage, compared to best-efforts NBN where there is no guaranteed restoration time.

The business internet SLA on a dedicated service does not prevent outages, but it contractually limits how long they last and often includes financial remedies if the SLA is breached. For businesses where the cost of downtime is very high and they cannot tolerate even short outages, a dedicated service with a strong SLA combined with a 4G failover for the rare cases that exceed the SLA is a robust combination.

Dedicated internet is typically more expensive than NBN business services, and it is most commonly used by larger businesses, those with high symmetrical bandwidth requirements, or those where data security and consistent performance are critical. The comparison between NBN business and enterprise ethernet covers these differences in more detail.

The 4G Failover Sweet Spot for Australian SMBs

For the majority of Australian small and medium businesses, the most cost-effective and practical redundancy solution is a quality business NBN plan with a 4G failover router. This combination addresses the primary failure scenarios — provider outages, cable faults, and equipment issues — at a total ongoing cost that is typically $50 to $80 per month above the base NBN plan cost, assuming a failover-capable router is already in place.

When you consider that a single half-day outage can cost a business more in lost productivity, missed sales, and frustrated customers than a full year of failover SIM costs, the value proposition is not difficult to make.

The practical coverage picture for 4G in Australia supports this approach. Telstra, Optus, and TPG/Vodafone all provide strong 4G coverage across metropolitan and most suburban areas. Telstra's 4G network extends to many regional areas where NBN is the primary connection technology and where the distance from a network exchange can make fault restoration times longer than average. For these businesses, 4G failover is not just cost-effective — it is often the only practical secondary option.

4G speeds during failover — typically 20 to 100 Mbps depending on location and tower load — are sufficient for the business functions that matter most during an outage: VoIP calls, accessing cloud-based software, processing EFTPOS transactions, receiving and sending email, and maintaining remote access for staff working from home or on the road.

Where 4G failover becomes insufficient is in businesses where the secondary connection needs to carry the same workload as the primary — for example, businesses with many concurrent video conference calls or those that regularly upload or download very large files. In these cases, a dual fixed WAN configuration or a higher-capacity 5G failover solution is more appropriate.

What Should Route Over the Failover Connection?

Not all internet traffic is equally critical during an outage, and a well-configured failover setup accounts for this. The goal is to ensure that revenue-generating and operationally essential traffic continues without interruption, while non-essential or bandwidth-heavy traffic is either deprioritised or temporarily suspended.

Traffic that should be prioritised over the failover connection includes VoIP calls and your business phone system, EFTPOS and payment terminal traffic, cloud-based business software such as accounting platforms and CRM systems, email and communication tools, and remote desktop or VPN access for staff who need to remain productive.

Traffic that can be deprioritised or suspended during failover includes software updates and patch downloads, cloud backup jobs that involve large data transfers, video streaming, and large file synchronisation with cloud storage platforms.

A properly configured business-grade router or firewall can handle this traffic prioritisation automatically using quality of service (QoS) rules. QoS rules can be configured to guarantee bandwidth for specific traffic types — VoIP packets, for example — and to throttle or block lower-priority traffic when the available connection speed falls below a defined threshold.

This configuration work is what separates a robust redundancy setup from a simple hardware purchase. The router with two WAN ports and a SIM slot is necessary, but it is not sufficient. The failover triggers, the QoS rules, and the traffic prioritisation configuration all need to be set up correctly and tested before they are needed.

Dual WAN Routers — What to Look For

If you are evaluating hardware for a dual WAN redundancy setup, there are several capabilities that distinguish business-grade equipment from consumer hardware that technically has a secondary WAN port but lacks the management features to use it reliably.

Automatic failover, not manual switchover, is the baseline requirement. The router should detect connection failure and redirect traffic without any staff intervention. The failover trigger should be configurable — the most reliable method is a combination of ping testing to multiple external hosts (such as 8.8.8.8 and 1.1.1.1) and DNS resolution testing, so that the router correctly distinguishes between "the connection is up but nothing is working" and "the connection is genuinely down."

Configurable failover detection intervals matter. A router that takes five minutes to detect a connection failure and then another two minutes to complete the switchover is far less useful than one that detects and switches in under 60 seconds. For VoIP applications in particular, a slow failover means dropped calls.

Load balancing capability allows traffic to be distributed across both connections during normal operation, which is a bonus capability beyond basic failover. VPN compatibility is important if your staff use remote access, as some dual WAN routers handle VPN traffic poorly when connections switch.

QoS controls allow you to define traffic priorities that apply during failover, ensuring that critical applications get bandwidth over non-critical ones.

For Australian SMBs, well-regarded options include the Draytek Vigor series, which has a long history of deployment in Australian business networks and strong local support; the Peplink Balance series, which is particularly suited to mobile-heavy environments; and Fortinet's small business FortiGate range, which adds firewall and security capabilities alongside dual WAN management. Your managed IT provider or your internet provider can advise on the appropriate model for your specific requirements.

Testing Your Failover — Don't Wait for a Real Outage

This point is worth emphasising clearly: a failover setup that has never been tested is an assumption, not a capability. Many businesses invest in dual WAN hardware, have it configured, and then discover during an actual outage that the failover does not work as expected — perhaps because the SIM has been deactivated due to inactivity, the failover trigger is misconfigured, or a firmware update changed a setting.

The only way to know your failover works is to test it deliberately, under controlled conditions, before you need it.

The test is straightforward. During business hours, when your network is in normal use, physically disconnect your primary NBN connection — unplug the cable between your NBN connection device and your router's primary WAN port. Then verify the following. The router detects the failure and switches to the secondary connection within the expected timeframe, typically under 60 seconds. Devices on the network can access the internet over the secondary connection without any manual configuration changes. VoIP calls can be made and received without reconnecting or re-registering. Cloud-based applications remain accessible. Payment terminals can process transactions.

After the test, reconnect the primary connection and verify that the router returns to primary operation automatically.

This test should be performed when the redundancy setup is first commissioned, and then repeated at least annually or after any significant changes to your network configuration — a router firmware update, a change of SIM provider, an upgrade to your NBN service, or the addition of new devices or applications to your network.

Documenting the test results — date, outcome, failover time, any issues identified — gives you a simple record that the system is working and provides a baseline for diagnosing problems if a real outage reveals unexpected behaviour.

How Pickle Supports Business Internet Redundancy

Pickle's business internet plans are designed with operational reliability as a priority, and we understand that for most Australian SMBs, the combination of a quality NBN business service and a properly configured 4G failover solution is the right starting point.

We work with Australian SMBs to design internet infrastructure that matches their actual risk profile and operational requirements. That means recommending the right NBN plan for your bandwidth needs, advising on hardware that supports genuine automatic failover, helping configure the failover triggers and QoS rules that make the difference between a theoretical and a functional redundancy setup, and providing the ongoing support that keeps everything working as your business changes.

If you have experienced an outage and are looking to ensure it does not happen again, or if you are reviewing your infrastructure proactively, we are happy to work through the options with you.

Call us on 1300 688 588 or email [email protected] to speak with the team.

Frequently Asked Questions

Q: How much does business internet redundancy cost in Australia?

A: For most Australian SMBs, the most accessible redundancy solution is a 4G failover router combined with a business data SIM. The hardware investment is typically $300 to $800 for a quality dual WAN router, and the ongoing SIM cost is $30 to $80 per month depending on the data allowance and carrier. If you already have a compatible router in place, the ongoing cost can be as low as $30 to $50 per month. A dual fixed WAN setup — running two fixed-line services simultaneously — typically costs $150 to $400 per month across both service fees. SD-WAN and dedicated enterprise internet solutions are priced higher and are generally appropriate for larger businesses or multi-site operations.

Q: Does 4G failover work automatically without staff intervention?

A: Yes, provided the router is configured correctly. A properly set up 4G failover router continuously monitors the primary connection using ping tests or DNS queries to external hosts. When the router detects that the primary connection has failed — typically within 30 to 60 seconds — it automatically routes all outbound traffic over the 4G connection. Staff do not need to do anything. When the primary connection restores, the router automatically returns to it. The key word is "configured correctly" — the failover triggers need to be set up and tested to ensure they behave as expected. A router that has been installed but never configured for automatic failover will not work.

Q: Will VoIP phones work over a 4G failover connection?

A: In most cases, yes. VoIP calls require relatively modest bandwidth — typically 100 kbps or less per call — and 4G connections in Australian metropolitan and suburban areas comfortably support this. The more important consideration for VoIP over 4G failover is latency and packet loss. 4G networks in well-covered areas typically deliver latency suitable for VoIP calls. Problems arise in areas with congested towers or weak signal, where latency and packet loss can affect call quality. A QoS configuration that prioritises VoIP traffic over the failover connection helps ensure that call quality is maintained even when other devices are also using the 4G connection. Testing VoIP specifically during your failover test is strongly recommended.

Q: Should I have two connections from different providers or just different technologies?

A: Ideally both. The purpose of using different technologies — for example, NBN fixed line and 4G mobile — is to ensure that the same physical infrastructure failure cannot disable both connections simultaneously. Two NBN services from different providers in the same area often share physical infrastructure (the copper or fibre to the street, the NBN Co exchange equipment) and may both fail during the same local fault. A 4G connection uses entirely separate infrastructure — mobile towers, a mobile carrier's core network — and is unaffected by NBN cable or exchange faults. Where possible, using different providers in addition to different technologies adds a further layer of independence, but the technology difference is more important than the provider difference.

Q: What is SD-WAN and do I need it for my small business?

A: SD-WAN (software-defined wide area network) is a technology platform that manages traffic across multiple internet connections simultaneously, routing different types of traffic over whichever connection is most appropriate at any given time based on performance, cost, or policy rules. It goes beyond simple failover into active, continuous traffic optimisation. For most single-site Australian SMBs, SD-WAN is more complexity and cost than necessary — a dual WAN router with automatic 4G failover achieves the primary goal of keeping the business online during an outage at a fraction of the cost. SD-WAN becomes relevant for businesses with multiple offices that need consistent application performance across sites, businesses with very high traffic volumes across mixed connection types, or organisations with specific compliance or security requirements that benefit from SD-WAN's policy controls. If you are not sure whether SD-WAN is appropriate for your business, the short answer for a single-site SMB is almost certainly no — start with a dual WAN router and 4G failover.